CN118097619B - Intelligent highway damage inspection method and system based on Internet of things - Google Patents
Intelligent highway damage inspection method and system based on Internet of things Download PDFInfo
- Publication number
- CN118097619B CN118097619B CN202410276473.8A CN202410276473A CN118097619B CN 118097619 B CN118097619 B CN 118097619B CN 202410276473 A CN202410276473 A CN 202410276473A CN 118097619 B CN118097619 B CN 118097619B
- Authority
- CN
- China
- Prior art keywords
- area
- crack
- determining
- blocking
- overall
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
- G06V10/225—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition based on a marking or identifier characterising the area
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The application relates to an intelligent highway damage inspection method and system based on the Internet of things, and relates to the field of highway engineering technology; determining a pavement area, a crack area and an abnormal area in the pavement acquisition image; dividing the abnormal region into a class region and a blocking region; determining the center position of the area according to the blocking area, generating a cleaning operation path according to the center positions and the initial positions of all the areas, and controlling the cleaning device to move along the cleaning operation path until the blocking area is completely determined as a crack area or a pavement area; and when the blocking area does not exist, generating a crack marking point according to the crack area and the real-time position of the device and uploading the crack marking point to a preset data processing center. The application has the effect of improving the overall operation effect when the damage inspection unmanned vehicle performs damage inspection operation on the highway.
Description
Technical Field
The application relates to the field of highway engineering technology, in particular to an intelligent highway damage inspection method and system based on the Internet of things.
Background
The intelligent road, also called intelligent traffic system, is a new road system integrating various advanced technologies such as internet of things, big data, cloud computing, artificial intelligence and the like. The intelligent highway is provided with an intelligent solution for traffic management and travel by collecting and analyzing the road service condition in real time through various sensor devices and communication devices.
In the related art, a method for inspecting the damage of an intelligent highway generally assigns a damaged inspection unmanned vehicle to move on the highway to obtain a surface image of the highway, and analyzes the image to determine whether a crack appears on the highway and the width and depth of the crack when the crack appears, thereby determining the damage information of the highway.
In view of the above-mentioned related art, the inventor believes that when an unmanned vehicle acquires an image of a road surface, if there is an external foreign matter (for example, garbage, sludge, etc.) in the image, the image of the actual surface of the road cannot be acquired, and thus the damage condition of the road cannot be effectively determined, resulting in poor damage inspection operation effect on the road, and there is still room for improvement.
Disclosure of Invention
In order to improve the overall operation effect when the damage inspection unmanned vehicle performs damage inspection operation on the highway, the application provides an intelligent highway damage inspection method and system based on the Internet of things.
In a first aspect, the application provides an intelligent highway damage inspection method based on the internet of things, which adopts the following technical scheme:
An intelligent highway damage inspection method based on the Internet of things comprises the following steps:
acquiring the real-time position of the device and a road surface acquisition image;
performing feature recognition in the road surface acquisition image to determine a road surface area and a non-road surface area, and performing feature recognition in the non-road surface area to determine a crack area and an abnormal area;
Determining an abnormal overall area according to the abnormal area, defining the abnormal area with the abnormal overall area smaller than a preset reference area as a similar area, and defining the rest abnormal areas as blocking areas;
judging whether the similar area is contacted with the crack area or not;
if the similar area is contacted with the crack area, determining the similar area as the crack area;
If the similar area is not contacted with the crack area, determining the similar area as a pavement area;
Determining the center position of the area according to the blocking area, generating a cleaning operation path according to the center positions of all the areas and a preset initial position, and controlling a preset cleaning device to move along the cleaning operation path;
re-acquiring an image of the blocking area after the cleaning device is moved, judging a crack area and a pavement area of the blocking area, and controlling the cleaning device to operate again when the blocking area still exists until the blocking area is completely determined as the crack area or the pavement area;
and when the blocking area does not exist, generating a crack marking point according to the crack area and the real-time position of the device and uploading the crack marking point to a preset data processing center.
Optionally, the step of determining the location of the center of the area based on the blocked area includes:
generating a surrounding area which can surround the blocking area and is rectangular in shape, and determining the surrounding area according to the surrounding area;
determining a surrounding area with the smallest numerical value according to a preset ordering rule, and determining a surrounding area corresponding to the surrounding area as a circumscribed area of the blocking area;
generating a simulation center point which can move at will in the circumscribed area, and determining the point spacing distance according to the simulation center point and each contour point of the blocking area;
Calculating the average value according to the distance between all the point positions to determine the distance between the average value and the distance between all the point positions to determine the whole deviation distance;
And determining the overall deviation distance with the smallest numerical value according to the ordering rule, and determining the simulation center point corresponding to the overall deviation distance as the region center position.
Optionally, after the overall deviation distance is determined, the intelligent highway damage inspection method based on the internet of things further includes:
judging whether at least two simulation center points with the same and minimum integral deviation distance values exist or not;
if at least two simulation center points with the same overall deviation distance value and the minimum value do not exist, determining the center position of the region according to the corresponding simulation center points;
if at least two simulation center points with the same and minimum overall deviation distance values exist, eliminating the point separation distance with the largest value and the smallest value in the point separation distances, and calculating the average separation distance again after eliminating;
Performing difference calculation according to the average value interval distance calculated successively to determine a rejection deviation value;
And determining a rejection deviation value with the largest value according to the sorting rule, and determining the center position of the region according to the simulation center point corresponding to the rejection deviation value.
Optionally, the step of generating the cleaning job path according to all the center positions of the areas and the preset initial position includes:
Combining the center positions of all the areas according to any number to determine a center position combination;
The central positions of all the areas in the central position combination are sequenced in any sequence and connected to determine the point position movement track;
Determining a cleaning area according to the point position moving track and a preset device processing range, defining the cleaning area containing all the blocking areas as a qualified area, and defining the point position moving track corresponding to the qualified area as a qualified moving track;
and determining the overall cleaning distance according to the qualified moving track, determining the overall cleaning distance with the minimum value according to the sorting rule, and determining the qualified moving track corresponding to the overall cleaning distance as a cleaning operation path.
Optionally, after the overall cleaning distance is determined, the intelligent highway damage inspection method based on the internet of things further includes:
judging whether at least two qualified moving tracks with the same and minimum overall cleaning distance values exist or not;
If at least two qualified moving tracks with the same and minimum overall cleaning distance values do not exist, determining a cleaning operation path according to the unique qualified moving tracks;
if at least two qualified moving tracks with the same and minimum overall cleaning distance values exist, determining a single body rotation angle according to the central position of each area in each qualified moving track;
Calculating according to all the single rotation angles to determine the integral rotation angle;
and determining the overall rotation angle with the minimum numerical value according to the sorting rule, and determining the corresponding qualified moving track as a cleaning operation path according to the overall rotation angle.
Optionally, when the blocking area still exists, the intelligent highway damage inspection method based on the internet of things further includes:
controlling the preset cleaning times with the initial value of zero to be added with one treatment and outputting the actual treatment times;
judging whether the actual processing times are larger than the preset upper limit times or not;
if the actual treatment times are not greater than the upper limit times, controlling the cleaning device to operate again;
If the actual processing times are greater than the upper limit times, marking the corresponding blocking area as an area incapable of being processed, and generating a crack marking point according to the determined crack area;
Determining an area which cannot be processed according to the area which cannot be processed, and counting according to the crack area which is contacted with the area which cannot be processed so as to determine the number of peripheral cracks;
and calculating according to the area which cannot be processed and the number of the peripheral cracks to determine the area theoretical crack ratio, and binding the corresponding area theoretical crack ratio with the area which cannot be processed to upload the area to a data processing center.
Optionally, after the theoretical crack ratio is determined, the intelligent highway damage inspection method based on the internet of things further includes:
calculating a difference value according to a preset image acquisition area and an area which cannot be processed so as to determine a successful identification area;
Determining the overall crack area according to the crack area, and calculating according to the overall crack area and the successful recognition area to determine the overall actual crack occupation ratio;
Calculating a difference value according to the integral actual crack ratio and the regional theoretical crack ratio to determine a crack ratio difference value;
And determining compensation parameters corresponding to the crack proportion difference value according to a preset compensation matching relation, and updating the theoretical crack proportion of the region according to the compensation parameters.
In a second aspect, the application provides an intelligent highway damage inspection system based on the Internet of things, which adopts the following technical scheme:
An intelligent highway damage inspection system based on the internet of things, comprising:
the acquisition module is used for acquiring the real-time position of the device and the road surface acquisition image;
The processing module is connected with the acquisition module and the judging module and is used for storing and processing information;
the judging module is connected with the acquisition module and the processing module and is used for judging information;
The processing module performs feature recognition in the road surface acquisition image to determine a road surface area and a non-road surface area, and performs feature recognition in the non-road surface area to determine a crack area and an abnormal area;
the processing module determines an abnormal overall area according to the abnormal area, defines the abnormal area with the abnormal overall area smaller than a preset reference area as a similar area, and defines the rest abnormal areas as blocking areas;
the judging module judges whether the class area is contacted with the crack area or not;
if the judging module judges that the similar area is contacted with the crack area, the processing module determines the similar area as the crack area;
If the judging module judges that the similar area is not contacted with the crack area, the processing module determines the similar area as a pavement area;
the processing module determines the center position of the area according to the blocking area, generates a cleaning operation path according to the center positions of all the areas and a preset initial position, and controls a preset cleaning device to move along the cleaning operation path;
The processing module acquires the image of the blocking area again after the cleaning device is moved, judges the crack area and the pavement area of the blocking area, and controls the cleaning device to operate again when the blocking area still exists until the blocking area is completely determined as the crack area or the pavement area;
and when the blocking area does not exist, the processing module generates a crack marking point according to the crack area and the real-time position of the device and uploads the crack marking point to a preset data processing center.
In summary, the present application includes at least one of the following beneficial technical effects:
When the damage inspection unmanned vehicle is used for inspecting cracks on the highway, the area with the external foreign matters can be determined, and the corresponding cleaning device is controlled to clean the external foreign matters, so that shielding of the external foreign matters on the surface of the highway is reduced, and the overall operation effect of the damage inspection unmanned vehicle when the damage inspection unmanned vehicle performs damage inspection operation on the highway is improved;
A more reasonable cleaning operation path can be set according to different position conditions of the blocking area in the image, so that the overall effect of the cleaning device for cleaning the external foreign matters is better;
the crack in the area shielded by the external foreign matters can be theoretically analyzed according to the overall crack condition aiming at the situation that the external foreign matters cannot be cleaned, so that corresponding data are provided for staff to analyze.
Drawings
Fig. 1 is a flowchart of an intelligent highway damage inspection method based on the internet of things.
Fig. 2 is a flow chart of a method of determining a location of a center of a region.
Fig. 3 is a flow chart of a simulated center point screening method.
FIG. 4 is a flow chart of a cleaning job path generation method.
Fig. 5 is a flowchart of a qualified movement trajectory screening method.
Fig. 6 is a flowchart of a method of predicting an area crack that cannot be processed.
Fig. 7 is a flowchart of a crack ratio correction method.
Fig. 8 is a block flow diagram of an intelligent highway damage inspection method based on the internet of things.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings 1 to 8 and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
Embodiments of the application are described in further detail below with reference to the drawings.
The embodiment of the application discloses an intelligent highway damage inspection method based on the Internet of things, which can be used for determining and analyzing external foreign matters shielding a highway when a highway damage unmanned vehicle is used for detecting the damage condition of the highway, and controlling a corresponding cleaning device to clean the external foreign matters, so that the surface of the highway can be effectively detected, and the overall operation effect of the highway damage inspection unmanned vehicle during the damage inspection operation is improved.
Referring to fig. 1, the method flow of the intelligent highway damage inspection method based on the internet of things comprises the following steps:
Step S100: the real-time position of the device and the road surface acquisition image are acquired.
The real-time position of the device is the position of the current damaged inspection unmanned vehicle, and the position can be obtained by installing a corresponding GPS on the unmanned vehicle; the road surface captured image is an image with a road surface obtained by a photographing device mounted on the unmanned vehicle, and the photographed position is determined specifically by the position and orientation in which the photographing device is mounted.
Step S101: feature recognition is performed in the road surface acquisition image to determine a road surface area and a non-road surface area, and feature recognition is performed in the non-road surface area to determine a crack area and an abnormal area.
The road surface area is an area of a road surface which is not cracked and can be effectively identified, the non-road surface area is an area of the road surface which is not a road surface area, the characteristic identification method can be that a plurality of images are acquired on the road surface which is normally free of external foreign matters and is not cracked according to a specific road to be inspected, a training road surface area identification model is learned through a neural network, and the road surface area and the non-road surface area can be determined by importing corresponding road surface acquisition images; the crack area is an area where cracking occurs on the surface of the highway, and the abnormal area is an area which cannot be effectively identified, for example, an area where soil and external garbage exist, and the specific characteristic identification method is the same as the above, and is not repeated.
Step S102: and determining an abnormal overall area according to the abnormal area, defining the abnormal area with the abnormal overall area smaller than a preset reference area as a class area, and defining the rest abnormal areas as a blocking area.
The abnormal whole area is the area of a single abnormal area, which is determined by the condition that the abnormal area occupies the pixels of the road surface collected image; the standard area is the maximum external foreign matter area that the staff set up when can not analyze the road surface crack, exists on the road surface for example earth, and self coverage is little to can not produce the influence to crack analysis, need not the processing that corresponds this moment, and the big piece rubbish that exists on the road surface, then the coverage is big, can produce the influence to crack analysis, defines class region and blocks the region in order to effectively distinguish different unusual regions, the follow-up analysis of being convenient for.
Step S103: and judging whether the similar area is contacted with the crack area or not.
The purpose of the judgment is to know whether a highway crack exists below the same area.
Step S1031: if the similar region is in contact with the fracture region, the similar region is determined to be the fracture region.
When the similar area is contacted with the crack area, the crack on the road extends to the similar area, namely, the crack exists below the foreign matter at the similar area, and the similar area is determined to be the crack area so as to effectively determine the road damage condition.
Step S1032: if the similar area is not in contact with the crack area, the similar area is determined as the pavement area.
When the similar area is not contacted with the crack area, the crack on the road is not extended to the similar area, namely the possibility that the crack exists below the foreign matter at the similar area is low, and the crack is determined as the road surface area at the moment so as to effectively determine the road damage condition.
Step S104: determining the center position of the area according to the blocking area, generating a cleaning operation path according to all the center positions of the area and a preset initial position, and controlling a preset cleaning device to move along the cleaning operation path.
The central position of the region is a central point position which can represent the center of the blocking region, in one embodiment, each contour point on the contour line of the blocking region can be connected to determine the length between each contour point, and then the central point of the two contour points which are farthest apart is taken as the central position of the region, in another embodiment, the central point can be determined according to the method from step S200 to step S204, and the description is omitted here; the initial position is a position point relative to the image when the cleaning device is not in operation, wherein the cleaning device is a device with a moving function and a rotating function, such as a rotating broom below a garbage truck; the cleaning operation path is a path moved when the cleaning device cleans the external foreign matters on the determined blocking area, in one embodiment, the central positions of all the areas can be sorted from left to right and from top to bottom, and then all the position points are connected one by one according to the sorting by taking the initial position as a starting point to determine the cleaning operation path, in another embodiment, the cleaning operation path can be determined according to the method from the step S400 to the step S403, and the description is omitted here; the control cleaning device moves along the cleaning operation path to clean the foreign matters at the blocking area, so that the specific power damage condition can be determined conveniently.
Step S105: and re-acquiring an image of the blocking area after the cleaning device is moved, judging the crack area and the pavement area of the blocking area, and controlling the cleaning device again when the blocking area still exists until the blocking area is completely determined as the crack area or the pavement area.
The foreign matters which can be processed can be effectively processed through continuous operation of the cleaning device, so that the type of the blocking area can be effectively determined; for the case that the foreign object cannot be handled, the following steps S600 to S603 are described, and will not be described here.
Step S106: and when the blocking area does not exist, generating a crack marking point according to the crack area and the real-time position of the device and uploading the crack marking point to a preset data processing center.
The real-time position of the device and the position point of the crack area in the image can be determined, so that the corresponding crack mark point can be uploaded to a data processing center, a worker can know the specific damage condition of the highway, and the subsequent maintenance operation is convenient.
Referring to fig. 2, the step of determining the region center position from the blocking region includes:
Step S200: a surrounding area which can surround the blocking area and is rectangular in shape is generated, and the surrounding area is determined according to the surrounding area.
The surrounding area is an area which can completely surround the blocking area, and one blocking area has countless surrounding areas which meet the requirements, namely the area of the surrounding area.
Step S201: and determining the surrounding area with the smallest numerical value according to a preset ordering rule, and determining the surrounding area corresponding to the surrounding area as the circumscribed area of the blocking area.
The sorting rule is a method which is set by staff and can sort the values, such as a bubbling method, and the bounding area with the smallest value can be determined through the sorting rule, namely, the bounding area corresponding to the bounding area just carries out bounding processing on the blocking area, and the bounding area is defined as an circumscribed area at the moment so as to facilitate subsequent processing analysis.
Step S202: and generating a simulation center point which can move at will in the circumscribed area, and determining the point spacing distance according to the simulation center point and each contour point of the blocking area.
The simulated center point is a movable point position at any position in the circumscribed area, and the distance between the point positions is the distance value between the simulated center point and the contour point of the blocking area.
Step S203: and carrying out average value calculation according to all the point separation distances to determine an average value separation distance, and carrying out calculation according to the average value separation distance and all the point separation distances to determine an overall deviation distance.
The average value interval is the average value of all the point separation distances determined by a single circumscribed area, and the integral deviation distance is the sum of absolute values of differences between the average value interval and the point separation distances.
Step S204: and determining the overall deviation distance with the smallest numerical value according to the ordering rule, and determining the simulation center point corresponding to the overall deviation distance as the region center position.
The overall deviation distance with the smallest value can be determined through the sorting rule, namely the distance deviation between the simulation center point and each contour point is the smallest at the moment, namely the simulation center point can be used as the center point of the blocking area, and accordingly the corresponding simulation center point can be determined to be the area center position.
Referring to fig. 3, after the overall deviation distance is determined, the intelligent highway damage inspection method based on the internet of things further includes:
Step S300: and judging whether at least two simulation center points with the same and minimum integral deviation distance values exist.
The purpose of the determination is to know whether there are multiple analog center points that meet the requirements.
Step S3001: and if at least two simulation center points with the same overall deviation distance value and the minimum value do not exist, determining the center position of the region according to the corresponding simulation center points.
When at least two simulation center points with the same overall deviation distance value and the minimum value do not exist, only the only simulation center point meeting the requirements is indicated, and the center position of the area is normally determined.
Step S3002: if at least two simulation center points with the same and minimum overall deviation distance values exist, eliminating the point separation distance with the largest value and the smallest value in the point separation distances, and calculating the average separation distance again after eliminating.
When at least two simulation center points with the same and minimum integral deviation distance values exist, the fact that a plurality of simulation center points meeting the requirements exist is indicated, and further screening is needed; and eliminating the point distance with the largest value and the smallest value in the point distance to reduce the influence caused by partial data with larger deviation quantity, thereby further analyzing the specific central point situation.
Step S301: and carrying out difference value calculation according to the average value interval distance calculated in sequence to determine a rejection deviation value.
The rejecting offset value is the difference between successively calculated mean separation distances, which is the absolute value.
Step S302: and determining a rejection deviation value with the largest value according to the sorting rule, and determining the center position of the region according to the simulation center point corresponding to the rejection deviation value.
The largest value of the eliminating deviation value can be determined through the sorting rule, namely the eliminating deviation value has the largest influence on the average value distance at the moment, namely the distance deviation of the simulation center point from other contour points is smaller, and the corresponding simulation center point can be determined as the center position of the region at the moment.
Referring to fig. 4, the step of generating a cleaning job path according to all the area center positions and the preset initial positions includes:
step S400: all of the zone center locations are arbitrarily combined according to any number to determine a center location combination.
The center position combinations are combinations formed according to the selected center positions of the areas, for example, if there are three center positions A, B, C of the areas, the corresponding center position combinations include A, B, C, AB, AC, BC, ABC, and seven kinds of center position combinations are used.
Step S401: and sequencing and connecting the central positions of all the areas in the central position combination in any order to determine the point position movement track.
The point location movement track is a track obtained by sequentially connecting the central positions of the areas according to the determined sequence by taking the initial position as a starting point, for example, the initial position is O, and taking the central position combination AB as an example, the obtained point location movement tracks are OAB and OBA respectively.
Step S402: and determining a cleaning area according to the point position moving track and a preset device processing range, defining the cleaning area containing all the blocking areas as a qualified area, and defining the point position moving track corresponding to the qualified area as a qualified moving track.
The device processing range is the range of the cleaning device for processing the foreign matters, for example, when the cleaning device is a rotary broom on a garbage truck, the corresponding device processing range is a circle formed by the rotary radius of the broom; the cleaning area is the area which can be cleaned by the cleaning device moving along the point location moving track, when all the blocking areas are included, the foreign matters on all the determined blocking areas can be cleaned, and the corresponding cleaning area is defined as a qualified area so as to realize the distinction of different cleaning areas; meanwhile, the corresponding point location moving track is defined as a qualified moving track so as to distinguish different point location moving tracks, and the cleaning operation path can be conveniently selected subsequently.
Step S403: and determining the overall cleaning distance according to the qualified moving track, determining the overall cleaning distance with the minimum value according to the sorting rule, and determining the qualified moving track corresponding to the overall cleaning distance as a cleaning operation path.
The whole cleaning distance is all distance values required to be moved when the cleaning device moves along the qualified moving track, and the whole cleaning distance with the minimum value can be determined through the sequencing rule, namely, the cleaning efficiency is fastest according to the qualified moving track operation corresponding to the whole cleaning distance, and the corresponding qualified moving track is determined to be the cleaning operation path.
Referring to fig. 5, after the overall cleaning distance is determined, the intelligent highway damage inspection method based on the internet of things further includes:
Step S500: judging whether at least two qualified moving tracks with the same and minimum overall cleaning distance values exist.
The purpose of the judgment is to know whether a plurality of qualified moving tracks meeting the requirements exist.
Step S5001: and if at least two qualified moving tracks with the same and minimum overall cleaning distance values do not exist, determining a cleaning operation path according to the unique qualified moving tracks.
When at least two qualified moving tracks with the same and minimum overall cleaning distance values do not exist, only the only qualified moving track meeting the requirements is indicated, and cleaning operation path determination is normally carried out at the moment.
Step S5002: if at least two qualified moving tracks with the same and minimum overall cleaning distance values exist, determining the single body rotation angle according to the central position of each area in each qualified moving track.
When at least two qualified moving tracks with the same and minimum overall cleaning distance values exist, a plurality of qualified moving tracks meeting the requirements are indicated to exist, and further screening treatment is needed; the single rotation angle is the angle of the cleaning device which needs to be moved and turned after moving to the central position of the area, namely the included angle formed by the moving path before the cleaning device moves to the central position of the area and the moving path after the cleaning device moves to the central position of the area.
Step S501: and calculating according to all the single rotation angles to determine the integral rotation angle.
The integral rotation angle is the sum of all the single rotation angles.
Step S502: and determining the overall rotation angle with the minimum numerical value according to the sorting rule, and determining the corresponding qualified moving track as a cleaning operation path according to the overall rotation angle.
The overall rotation angle with the minimum numerical value can be determined through the sorting rule, namely the rotation angle which needs to be adjusted to be minimum when the cleaning device moves along the corresponding qualified movement track is determined to be the cleaning operation path, so that the overall cleaning efficiency is improved.
Referring to fig. 6, when the blocking area still exists, the intelligent highway damage inspection method based on the internet of things further includes:
Step S600: and controlling the preset cleaning times with the initial value of zero to be added with one process and outputting the actual processing times.
The initial value of the cleaning times is zero, and when the moving operation of the cleaning device is completed, the cleaning times are added with one treatment, so that the actual treatment times, namely the total times of the moving operation of the current cleaning device, can be output.
Step S601: and judging whether the actual processing times are larger than the preset upper limit times.
The upper limit number of times is the minimum actual treatment number of times when the external foreign matters in the identification blocking area set by the staff cannot be removed by the cleaning device, and the purpose of judgment is to know whether the current external foreign matters can be removed by the cleaning device.
Step S6011: and if the actual processing times are not greater than the upper limit times, controlling the cleaning device to operate again.
When the actual treatment times are not more than the upper limit times, the possibility that the external foreign matters are cleaned still exists is indicated, and the cleaning device is controlled normally to perform cleaning operation again.
Step S6012: if the actual processing times are greater than the upper limit times, marking the corresponding blocking area as an area which cannot be processed, and generating a crack marking point according to the determined crack area.
When the actual treatment times are larger than the upper limit times, the possibility that the external foreign matters are cleaned by the cleaning device is low, and further analysis is needed; and defining the areas which cannot be processed to distinguish different areas, and simultaneously producing corresponding crack marking points so as to be convenient to upload to a data processing center.
Step S602: an area of inability to process is determined based on the area of inability to process, and the number of perimeter cracks is determined based on the area of cracks in contact with the area of inability to process.
The non-treatable area is the area of a single non-treatable area and the number of peripheral slits is the total number of slit areas in contact with the non-treatable area.
Step S603: and calculating according to the area which cannot be processed and the number of the peripheral cracks to determine the area theoretical crack ratio, and binding the corresponding area theoretical crack ratio with the area which cannot be processed to upload the area to a data processing center.
The theoretical crack ratio of the area is the ratio of the crack which can appear in the area which cannot be processed to the whole area which cannot be processed under the theoretical condition, and the calculation formula is thatWhereinCalculating coefficients for the theoretical crack occupation ratio of the region, wherein a is a preset formula, I is the number of peripheral cracks, and S is the area which cannot be processed; binding the theoretical crack proportion of the corresponding region with the region which cannot be processed and uploading the theoretical crack proportion of the corresponding region to a data processing center so as to provide support data for maintenance personnel.
Referring to fig. 7, after the theoretical crack ratio is determined, the intelligent highway damage inspection method based on the internet of things further includes:
step S700: and carrying out difference calculation according to the preset image acquisition area and the area which cannot be processed so as to determine the successful identification area.
The image acquisition area is the image area corresponding to the acquired road surface acquisition image, the successful identification area is the area of other areas except the area which cannot be processed in the image, and the area which cannot be processed is subtracted from the image acquisition area to determine.
Step S701: and determining the overall crack area according to the crack area, and calculating according to the overall crack area and the successfully identified area to determine the overall actual crack occupation ratio.
The total crack area is the sum of the areas of all the crack areas, the total actual crack ratio is the ratio of the cracks of other areas except the area which cannot be processed in the image, and the total crack area is divided by the successful identification area to determine.
Step S702: and calculating a difference value according to the integral actual crack ratio and the regional theoretical crack ratio to determine a crack ratio difference value.
The crack ratio difference is the overall actual crack ratio minus the regional theoretical crack ratio.
Step S703: and determining compensation parameters corresponding to the crack proportion difference value according to a preset compensation matching relation, and updating the theoretical crack proportion of the region according to the compensation parameters.
The compensation parameters are parameters for carrying out data correction on the regional theoretical crack ratio, when the crack ratio difference value is smaller, the determined regional theoretical crack ratio is higher, the corresponding compensation parameters are smaller and larger, the compensation matching relationship between the two is determined by a worker through multiple experiments in advance, and the regional theoretical crack ratio is updated by adding the original regional theoretical crack ratio to the compensation parameters, so that the determined regional theoretical crack ratio is more close to the actual condition, and the data real reliability is improved.
Referring to fig. 8, based on the same inventive concept, an embodiment of the present invention provides an intelligent road damage inspection method based on the internet of things, including:
the acquisition module is used for acquiring the real-time position of the device and the road surface acquisition image;
The processing module is connected with the acquisition module and the judging module and is used for storing and processing information;
the judging module is connected with the acquisition module and the processing module and is used for judging information;
The processing module performs feature recognition in the road surface acquisition image to determine a road surface area and a non-road surface area, and performs feature recognition in the non-road surface area to determine a crack area and an abnormal area;
the processing module determines an abnormal overall area according to the abnormal area, defines the abnormal area with the abnormal overall area smaller than a preset reference area as a similar area, and defines the rest abnormal areas as blocking areas;
the judging module judges whether the class area is contacted with the crack area or not;
if the judging module judges that the similar area is contacted with the crack area, the processing module determines the similar area as the crack area;
If the judging module judges that the similar area is not contacted with the crack area, the processing module determines the similar area as a pavement area;
the processing module determines the center position of the area according to the blocking area, generates a cleaning operation path according to the center positions of all the areas and a preset initial position, and controls a preset cleaning device to move along the cleaning operation path;
The processing module acquires the image of the blocking area again after the cleaning device is moved, judges the crack area and the pavement area of the blocking area, and controls the cleaning device to operate again when the blocking area still exists until the blocking area is completely determined as the crack area or the pavement area;
when the blocking area does not exist, the processing module generates a crack mark point according to the crack area and the real-time position of the device and uploads the crack mark point to a preset data processing center;
the area center position determining module is used for determining the area center position of each blocking area;
the simulation center point screening module is used for screening a plurality of simulation center points meeting the requirements to determine the unique region center position;
the cleaning operation path generation module is used for generating a proper cleaning operation path according to the central position of each area so as to enable the cleaning device to move;
The qualified moving track screening module is used for screening a plurality of qualified moving tracks meeting the requirements to determine a unique cleaning operation path;
the crack prediction module is used for predicting the crack condition of the area incapable of processing the external foreign matters;
and the crack ratio correction module is used for correcting the predicted crack ratio according to the overall crack condition of the area.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional modules is illustrated, and in practical application, the above-described functional allocation may be performed by different functional modules according to needs, i.e. the internal structure of the apparatus is divided into different functional modules to perform all or part of the functions described above. The specific working processes of the above-described systems, devices and units may refer to the corresponding processes in the foregoing method embodiments, which are not described herein.
Claims (5)
1. An intelligent highway damage inspection method based on the Internet of things is characterized by comprising the following steps:
acquiring the real-time position of the device and a road surface acquisition image;
performing feature recognition in the road surface acquisition image to determine a road surface area and a non-road surface area, and performing feature recognition in the non-road surface area to determine a crack area and an abnormal area;
Determining an abnormal overall area according to the abnormal area, defining the abnormal area with the abnormal overall area smaller than a preset reference area as a similar area, and defining the rest abnormal areas as blocking areas;
judging whether the similar area is contacted with the crack area or not;
if the similar area is contacted with the crack area, determining the similar area as the crack area;
If the similar area is not contacted with the crack area, determining the similar area as a pavement area;
Determining the center position of the area according to the blocking area, generating a cleaning operation path according to the center positions of all the areas and a preset initial position, and controlling a preset cleaning device to move along the cleaning operation path;
re-acquiring an image of the blocking area after the cleaning device is moved, judging a crack area and a pavement area of the blocking area, and controlling the cleaning device to operate again when the blocking area still exists until the blocking area is completely determined as the crack area or the pavement area;
Generating a crack mark point according to the crack region and the real-time position of the device when the blocking region does not exist, and uploading the crack mark point to a preset data processing center;
The step of determining the location of the center of the area from the blocked area comprises:
generating a surrounding area which can surround the blocking area and is rectangular in shape, and determining the surrounding area according to the surrounding area;
determining a surrounding area with the smallest numerical value according to a preset ordering rule, and determining a surrounding area corresponding to the surrounding area as a circumscribed area of the blocking area;
generating a simulation center point which can move at will in the circumscribed area, and determining the point spacing distance according to the simulation center point and each contour point of the blocking area;
Calculating the average value according to the distance between all the point positions to determine the distance between the average value and the distance between all the point positions to determine the whole deviation distance;
determining the overall deviation distance with the smallest value according to the ordering rule, and determining the simulation center point corresponding to the overall deviation distance as the center position of the region;
When the blocking area still exists, the intelligent highway damage inspection method based on the Internet of things further comprises the following steps:
controlling the preset cleaning times with the initial value of zero to be added with one treatment and outputting the actual treatment times;
judging whether the actual processing times are larger than the preset upper limit times or not;
if the actual treatment times are not greater than the upper limit times, controlling the cleaning device to operate again;
If the actual processing times are greater than the upper limit times, marking the corresponding blocking area as an area incapable of being processed, and generating a crack marking point according to the determined crack area;
Determining an area which cannot be processed according to the area which cannot be processed, and counting according to the crack area which is contacted with the area which cannot be processed so as to determine the number of peripheral cracks;
calculating according to the area incapable of being processed and the number of peripheral cracks to determine the area theoretical crack ratio, and binding the corresponding area theoretical crack ratio with the area incapable of being processed to upload to a data processing center;
After the theoretical crack ratio is determined, the intelligent highway damage inspection method based on the Internet of things further comprises the following steps:
calculating a difference value according to a preset image acquisition area and an area which cannot be processed so as to determine a successful identification area;
Determining the overall crack area according to the crack area, and calculating according to the overall crack area and the successful recognition area to determine the overall actual crack occupation ratio;
Calculating a difference value according to the integral actual crack ratio and the regional theoretical crack ratio to determine a crack ratio difference value;
And determining compensation parameters corresponding to the crack proportion difference value according to a preset compensation matching relation, and updating the theoretical crack proportion of the region according to the compensation parameters.
2. The internet of things-based intelligent road damage inspection method according to claim 1, wherein after the overall deviation distance is determined, the internet of things-based intelligent road damage inspection method further comprises:
judging whether at least two simulation center points with the same and minimum integral deviation distance values exist or not;
if at least two simulation center points with the same overall deviation distance value and the minimum value do not exist, determining the center position of the region according to the corresponding simulation center points;
if at least two simulation center points with the same and minimum overall deviation distance values exist, eliminating the point separation distance with the largest value and the smallest value in the point separation distances, and calculating the average separation distance again after eliminating;
Performing difference calculation according to the average value interval distance calculated successively to determine a rejection deviation value;
And determining a rejection deviation value with the largest value according to the sorting rule, and determining the center position of the region according to the simulation center point corresponding to the rejection deviation value.
3. The internet of things-based intelligent highway damage inspection method according to claim 1, wherein the step of generating the cleaning operation path according to all the area center positions and the preset initial position comprises:
Combining the center positions of all the areas according to any number to determine a center position combination;
The central positions of all the areas in the central position combination are sequenced in any sequence and connected to determine the point position movement track;
Determining a cleaning area according to the point position moving track and a preset device processing range, defining the cleaning area containing all the blocking areas as a qualified area, and defining the point position moving track corresponding to the qualified area as a qualified moving track;
and determining the overall cleaning distance according to the qualified moving track, determining the overall cleaning distance with the minimum value according to the sorting rule, and determining the qualified moving track corresponding to the overall cleaning distance as a cleaning operation path.
4. The intelligent road damage inspection method based on the internet of things according to claim 3, wherein after the overall cleaning distance is determined, the intelligent road damage inspection method based on the internet of things further comprises:
judging whether at least two qualified moving tracks with the same and minimum overall cleaning distance values exist or not;
If at least two qualified moving tracks with the same and minimum overall cleaning distance values do not exist, determining a cleaning operation path according to the unique qualified moving tracks;
if at least two qualified moving tracks with the same and minimum overall cleaning distance values exist, determining a single body rotation angle according to the central position of each area in each qualified moving track;
Calculating according to all the single rotation angles to determine the integral rotation angle;
and determining the overall rotation angle with the minimum numerical value according to the sorting rule, and determining the corresponding qualified moving track as a cleaning operation path according to the overall rotation angle.
5. Wisdom highway damage inspection system based on thing networking, its characterized in that includes:
the acquisition module is used for acquiring the real-time position of the device and the road surface acquisition image;
The processing module is connected with the acquisition module and the judging module and is used for storing and processing information;
the judging module is connected with the acquisition module and the processing module and is used for judging information;
The processing module performs feature recognition in the road surface acquisition image to determine a road surface area and a non-road surface area, and performs feature recognition in the non-road surface area to determine a crack area and an abnormal area;
the processing module determines an abnormal overall area according to the abnormal area, defines the abnormal area with the abnormal overall area smaller than a preset reference area as a similar area, and defines the rest abnormal areas as blocking areas;
the judging module judges whether the class area is contacted with the crack area or not;
if the judging module judges that the similar area is contacted with the crack area, the processing module determines the similar area as the crack area;
If the judging module judges that the similar area is not contacted with the crack area, the processing module determines the similar area as a pavement area;
the processing module determines the center position of the area according to the blocking area, generates a cleaning operation path according to the center positions of all the areas and a preset initial position, and controls a preset cleaning device to move along the cleaning operation path;
The processing module acquires the image of the blocking area again after the cleaning device is moved, judges the crack area and the pavement area of the blocking area, and controls the cleaning device to operate again when the blocking area still exists until the blocking area is completely determined as the crack area or the pavement area;
when the blocking area does not exist, the processing module generates a crack mark point according to the crack area and the real-time position of the device and uploads the crack mark point to a preset data processing center;
The processing module determines the central position of the area according to the blocking area, and comprises the steps of generating an enclosing area which can enclose the blocking area and is rectangular in shape by the processing module, and determining the enclosing area according to the enclosing area;
the processing module determines a surrounding area with the smallest numerical value according to a preset ordering rule, and determines a surrounding area corresponding to the surrounding area as a circumscribed area of the blocking area;
the processing module generates a simulation center point which can move at will in the circumscribed area, and determines the point spacing distance according to the simulation center point and each contour point of the blocking area;
The processing module calculates the average value according to the distance between all the point positions to determine the average value distance between the point positions, and calculates the average value distance between the point positions and the distance between all the point positions to determine the whole deviation distance;
The processing module determines the overall deviation distance with the smallest value according to the ordering rule, and determines the simulation center point corresponding to the overall deviation distance as the region center position;
When the blocking area still exists, the processing module controls the preset cleaning times with the initial value of zero to be added with one process and outputs the actual processing times;
The judging module judges whether the actual processing times are larger than the preset upper limit times or not;
if the judging module judges that the actual processing times are not more than the upper limit times, the processing module controls the cleaning device to operate again;
If the judging module judges that the actual processing times are greater than the upper limit times, the processing module marks the corresponding blocking area as an area which cannot be processed, and generates a crack mark point according to the determined crack area;
The processing module determines an area which cannot be processed according to the area which cannot be processed, and counts the number of peripheral cracks according to the crack area which is in contact with the area which cannot be processed;
the processing module calculates according to the area incapable of being processed and the number of the peripheral cracks to determine the area theoretical crack ratio, and binds the corresponding area theoretical crack ratio with the area incapable of being processed to upload the area incapable of being processed to the data processing center;
The theoretical crack occupation ratio determining post-processing module calculates a difference value according to a preset image acquisition area and an area which cannot be processed so as to determine a successful identification area;
The processing module determines the overall crack area according to the crack area, and calculates according to the overall crack area and the successful recognition area to determine the overall actual crack occupation ratio;
The processing module calculates the difference value according to the integral actual crack ratio and the regional theoretical crack ratio to determine the crack ratio difference value;
the processing module determines compensation parameters corresponding to the crack occupation ratio difference according to a preset compensation matching relation, and updates the theoretical crack occupation ratio of the area according to the compensation parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410276473.8A CN118097619B (en) | 2024-03-12 | 2024-03-12 | Intelligent highway damage inspection method and system based on Internet of things |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410276473.8A CN118097619B (en) | 2024-03-12 | 2024-03-12 | Intelligent highway damage inspection method and system based on Internet of things |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118097619A CN118097619A (en) | 2024-05-28 |
| CN118097619B true CN118097619B (en) | 2024-08-06 |
Family
ID=91159798
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410276473.8A Active CN118097619B (en) | 2024-03-12 | 2024-03-12 | Intelligent highway damage inspection method and system based on Internet of things |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118097619B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114973206A (en) * | 2022-07-28 | 2022-08-30 | 临沂市公路事业发展中心 | Automatic pavement disease identification and management method |
| CN115205819A (en) * | 2022-07-04 | 2022-10-18 | 成都霸天虎交通设施有限公司 | Municipal road surface sanitation cleaning intelligent analysis and regulation system based on machine vision |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4230395B2 (en) * | 2004-03-24 | 2009-02-25 | 三菱電機株式会社 | Surveillance mobile |
| US20230088335A1 (en) * | 2021-09-22 | 2023-03-23 | Nec Laboratories America, Inc. | Road and infrastructure analysis tool |
| KR102613308B1 (en) * | 2023-04-27 | 2023-12-13 | 노왕건 | Road management system using artificial intelligence |
| CN117474870B (en) * | 2023-11-03 | 2024-05-07 | 中国公路工程咨询集团有限公司 | Road pavement crack identification decision-making method and system based on big data screening |
-
2024
- 2024-03-12 CN CN202410276473.8A patent/CN118097619B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115205819A (en) * | 2022-07-04 | 2022-10-18 | 成都霸天虎交通设施有限公司 | Municipal road surface sanitation cleaning intelligent analysis and regulation system based on machine vision |
| CN114973206A (en) * | 2022-07-28 | 2022-08-30 | 临沂市公路事业发展中心 | Automatic pavement disease identification and management method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118097619A (en) | 2024-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111897318B (en) | Unmanned ship-based intelligent water pollution investigation method and device | |
| CN115268443B (en) | Robot obstacle avoidance path planning method | |
| CN112862704B (en) | A glue spraying and glue spraying quality inspection system based on 3D vision | |
| CN115167408B (en) | Routing inspection route planning system based on photovoltaic power station | |
| CN112557598A (en) | Urban air quality monitoring and management method based on Internet of things | |
| CN114814877B (en) | Tunnel data acquisition method, equipment and medium based on inspection robot | |
| CN111754638A (en) | Automatic dust suppression and dust fall system in storage yard and dust suppression and dust fall method in storage yard | |
| CN117809297B (en) | Three-dimensional reconstruction-based intelligent identification method for dangerous source of power transmission line | |
| CN116797055A (en) | Urban road planning method and system based on Internet of things | |
| CN117311395B (en) | Unmanned aerial vehicle path planning method and system for pollution inspection in river channel | |
| CN115941529B (en) | Cable tunnel detection method and system based on robot | |
| CN115880296A (en) | Machine vision-based prefabricated part quality detection method and device | |
| CN116758762A (en) | Control method based on big data | |
| CN116901089B (en) | Multi-angle vision distance robot control method and system | |
| CN112508911A (en) | Rail joint touch net suspension support component crack detection system based on inspection robot and detection method thereof | |
| CN116231504A (en) | Remote intelligent inspection method, device and system for booster station | |
| CN117669836A (en) | Atmospheric environment intelligent management and control method, system and storage medium | |
| CN118097619B (en) | Intelligent highway damage inspection method and system based on Internet of things | |
| CN117610972A (en) | Green building digital management system and method based on artificial intelligence | |
| CN107644533B (en) | Method for monitoring traffic flow of virtual section of expressway based on mobile network data | |
| CN114999161B (en) | Be used for intelligent traffic jam edge management system | |
| CN111862023A (en) | Pantograph structure abnormity detection method based on computer vision | |
| CN114581278A (en) | Method, terminal and system for traceability and investigation of air pollutants based on smart street lamps | |
| CN117875556A (en) | High-accuracy AI pipeline defect degree quantitative analysis system and method | |
| CN117726959A (en) | Unmanned aerial vehicle power line safety inspection system and method based on intelligent image recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |